Exploration of the mechanistic relationship between improved regional North
American inverse carbon fluxes and climate variability/trends

Principle Investigator: Kevin Gurney, Purdue University, kgurney@purdue.edu

Abstract:: We will use improved time-dependent net carbon exchange results from the TransCom intercomparison to advance quantitative understanding of the feedbacks between net carbon exchange and climate variability/change for all of the NICCR regions (RFP Focus 3). We improve upon recent inverse work through expansion and extension of observations, a state-of-the-art fossil fuel CO2 flux product, and sensitivity to interannually varying transport. The regional carbon-climate relationships will be further explored mechanistically through process datasets (NDVI, fire, drought) and terrestrial biosphere model results (focus 4 of RFP).

Location: All research will be performed at the Purdue campus, West Lafayette, Indiana USA.

Hypotheses: Primary Hypothesis: Can analysis of improved time-dependent North American inverse carbon exchange results yield mechanistic relationships to climate variability/change?
Sub-hypotheses:
1) Is the interannual variability/trends in NICCR region net carbon exchange related to leading modes of climate variability/trends (ENSO, AO, etc.) and/or observationally-based climate datasets (T, PPT, insolation, etc)? Are the lagged in time or teleconnected? 2) Can these relationships be quantified mechanistically through comparison to observationallybased process datasets such as NDVI, fire frequency/intensity, drought, eddy flux measurements and TBM output?

Methods: We will rerun the TransCom inverse experiment with a longer time period, multiple, expanded CO2 observation networks, and new N. Am. fossil fuel CO2 flux product (“Vulcan”).
• Three of the TransCom models will be rerun with the improved inverse setup and downscaled to estimate fluxes for the NICCR regions and subdivided boreal NA regions.
• These results will be related to observationally-based climate datasets (variables: temperature, precipitation, insolation, etc; indices: ENSO, PDO, etc) via lagged time series correlation, composite analysis, and spatial correlations.
• The resulting relationships will be mechanistically explored through relational analysis to observationally-based process datasets such as fire frequency/intensity, NDVI, drought and eddy flux measurements.
• The resulting relationships will be further explored through comparison to terrestrial biosphere model (TBM) results over the historical time period. This will identify integrated mechanistic relationships and offer the possibility of TBM improvement.

Deliverables: 1) Publicly available database of complete flux results, correlations (spatial and non-spatial), multivariate regression, composite analysis and error statistics. 2) Publicly available database of climate feedback indices by transport model, season, decade, and process. 3) Publication of results include: a. focus on temporal relationships to observationally-based climate and process datasets; b. focus on spatiotemporal teleconnections; c. focus on inter-comparison to TBM output and model improvement.



last updated: 28 February 2008 PLH